v0.0.16

What's Changed

• Add checkpoint loading warning by @WeberJulian in #73
• Custom callback by @erogol in #74

Full Changelog: v0.0.15...v0.0.16

v0.0.15

What's Changed

• Restore the best loss just for --continue_path by @Edresson in #67
• Print more info about training environment by @erogol in #66

Full Changelog: v0.0.14...v0.0.15

v0.0.14

What's Changed

• Fix Rank Zero Only Behavior by @iprovalo in #58

Full Changelog: v0.0.13...v0.0.14

v0.0.13

What's Changed

• Num gpus fix by @iprovalo in #47
• Profiling by @erogol in #49
• fix for loss not present in avg val by @manmay-nakhashi in #50
• Log training script by @erogol in #57

New Contributors

• @iprovalo made their first contribution in #47
• @manmay-nakhashi made their first contribution in #50

Full Changelog: v0.0.12...v0.0.13

v0.0.12

What's Changed

• Simple fit function for torch profiling by @erogol in #44

Full Changelog: v0.0.11...v0.0.12

v0.0.11

What's Changed

• Update for p3.10 by @erogol in #38

Full Changelog: v0.0.10...v0.0.11

v0.0.10

Just to fix the mismatch to the PyPI release version

v0.0.9

What's Changed

• Windows Multiprocessing Fix by @bitnom in #32
• Implement setup_training_environment by @erogol in #36

New Contributors

• @bitnom made their first contribution in #32

Full Changelog: v0.0.8...v0.0.9

v0.0.8

Full Changelog: v0.0.7...v0.0.8

• Set default cudnn_deterministic False

v0.0.7

What's Changed

• Add small run arg by @WeberJulian in #24
• Fix logger use by @erogol in #30

Full Changelog: v0.0.6...v0.0.7